ECG Artifact Removal from Single-Channel Surface EMG Using Fully Convolutional Networks

Enhanced ECG Record Quality: Integrated Artifact Suppression Using Soft Threshold on Wavelet Coefficients and Adaptive Filter Model

“…Furthermore, each electrode is also affected by other biomedical signal components during signal acquisition, such as electromyogram, electroencephalogram, and Electrooculogram..., especially the motion artifacts that occurs with voluntary or involuntary patient movement during ECG recording. Removal algorithm (QRSMR) [9]; Stationary Wavelet Movement Artifact Reduction (SWMAR); Normalized Least Mean Square Adaptive Filter technique (NLMSAF) [10]; moving average filtering, and wavelet transform have been used to reduce the motion ECG artefact [11,12]; removing such ECG artifacts from local field potentials (LFPs) recorded by a sensing-enabled neurostimulato [13]; ECG Artifact Removal from Single-Channel Surface EMG Using Fully Convolutional Networks [14], Channel-Wise Average Pooling and 1D Pixel-Shuffle Denoising Autoencoder for Electrode Motion Artifact Removal in ECG [15], Removing Cardiac Artifacts From Single-Channel Respiratory Electromyograms [16].…”

“…Figure 1. Amplitude-frequency distribution graph in ECG recording signal Many studies have focused on eliminating artifact on ECG recording signals with different approaches such as Motion artifact removal (MR); QRS detection based Motion ArtifactRemoval algorithm (QRSMR)[9]; Stationary Wavelet Movement Artifact Reduction (SWMAR); Normalized Least Mean Square Adaptive Filter technique (NLMSAF)[10]; moving average filtering, and wavelet transform have been used to reduce the motion ECG artefact[11,12]; removing such ECG artifacts from local field potentials (LFPs) recorded by a sensing-enabled neurostimulato[13]; ECG Artifact Removal from Single-Channel Surface EMG Using Fully Convolutional Networks[14], Channel-Wise Average Pooling and 1D Pixel-Shuffle Denoising Autoencoder for Electrode Motion Artifact Removal in ECG[15], Removing Cardiac Artifacts From Single-Channel Respiratory Electromyograms[16].The ECG signal recorded at the electrodes is a complex mixture of many components that come from different sources and are difficult to isolate. Using Independent Component Analysis (ICA) can help separate these sources into independent components (ICs), making it possible to remove unwanted components.However, the accuracy of ICA is highly dependent on the size of the analytical database[16], usually the number of signal sources in the body always exceeds the number of data recording channels; and in this case, ICA will not be able to separate the interference from the remaining components, or the components that are considered as artifacts, when removed still contain useful information, so the artifact cancellation will…”